This is a repository copy of Territorial Division in the Alfred-Guðrum Treaty : A Ninth Century Diplomatic Innovation?.
White Rose Research Online URL for this paper: https://eprints.whiterose.ac.uk/165125/
Version: Published Version
Article: Ashby, Steve orcid.org/0000-0003-1420-2108 and Marriott, Andrew (2020) Territorial Division in the Alfred-Guðrum Treaty : A Ninth Century Diplomatic Innovation? Apardjon Journal for Scandinavian Studies. pp. 22-53. ISSN 2634-0569
Reuse This article is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. This licence only allows you to download this work and share it with others as long as you credit the authors, but you can’t change the article in any way or use it commercially. More information and the full terms of the licence here: https://creativecommons.org/licenses/
Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [email protected] including the URL of the record and the reason for the withdrawal request.
[email protected] https://eprints.whiterose.ac.uk/ VOLUME 1
Apardjón Journal for Scandinavian Studies University of Aberdeen, UK 2020 Copyright © 2020 Steven P. Ashby, Jim Gritton, Sarah Künzler, Andrew G. Marriott, Blake Middleton, Ralph O’Connor, Roberto Luigi Pagani, Derek Parrott, Lyonel D. Perabo.
The authors published in this issue retain copyright of their submitted work and have granted Apardjón right of first publication of the work. This work is licensed under a Creative Commons Attribution-NonCommercial- NoDerivatives 4.0 International License. For more information, please visit creativecommons.org.
First published in 2020 ISSN 2634-0577 (Online) ISSN 2634-0569 (Print) Published in Aberdeen, United Kingdom
Apardjón Journal for Scandinavian Studies Centre for Scandinavian Studies University of Aberdeen Email: [email protected]
Main editors: Daniel Cutts, Heidi Synnøve Djuve, Deniz Cem Gülen, Ingrid Hegland, Jennifer Hemphill, Solveig Marie Wang, Caroline Wilhelmsson.
Cover image: © 2020 Marianne Mathieu Cover design and layout: Heidi Synnøve Djuve
The present volume of this periodical was financially supported by The Viking Society for Northern Research and the University of Aberdeen Development Trust Experience Fund. Territorial Division in the Alfred-Guðrum Treaty
– A Ninth-Century Diplomatic Innovation?
Andrew G. Marriott and Steven P. Ashby
Đis is ðæt frið, ðæt Ælfred cyninc ₇ Gyðrum cyning ₇ ealles Angelcynnes witan ₇ eal seo ðeod ðe on Eastæglum beoð ealle gecweden habbað ₇ mid aðum gefeostnod for hy sylfe ₇ for heora gingran, ge for geborene ge for ungeborene, ðe Godes miltse recce oððe ure (University of London/UCL 2019).
This is the peace which King Alfred and King Guthrum and the councillors of all the English race and all the people which is in East Anglia have all agreed on and confirmed with oaths, for themselves and for their subjects, both for the living and those yet unborn, who care to have God’s grace or ours (Whitelock 1979).
he study of past conflict and political accommodation is fundamental to archaeology, underlying any discussion of social dynamics, economics or identity. T The Treaty of Wedmore, from Viking-Age Britain, provides a key case study, with its division of Anglo-Saxon England between Scandinavian entities in the North and East and an Anglo-Saxon South. This phenomenon holds particular relevance today, as contemporary geopolitics increasingly demonstrates the need to appreciate the diversity, complexity and mutability of frontiers, as well as their relationships with military centres of gravity and buffer zones as, for example, in Northern Iraq, Syria, Libya, the Sahel and Nigeria. In such complex contestations of state or nationhood, the relationship between military occupation and early medieval state formation holds particular contemporary resonance. The Treaty of Wedmore was a significant outcome of Viking/Anglo-Saxon conflict in the ninth century (see Haslam 2006, 2012).1 The Treaty divided England into two distinct areas: King Alfred secured Wessex and much of Mercia, while East Anglia was given over to his Scandinavian adversary, King Guðrum. Though the area to the north and east of the border is often described as the Danelaw, that term has a later, eleventh-century provenance,
22 referring as much to the legal and administrative differences existing in those parts of the English kingdom that had, at various times, been under Viking control (Stenton 1989: 505- 507). Indeed, there is reason to see the Wedmore boundary as rather more ephemeral (e.g. Kershaw 2000), and it is certainly better understood within its contemporary context than as some more general marker of ‘Viking-Age’ territoriality, which was clearly a dynamic and mutable concept (Baker and Brookes 2012; Brookes and Reynolds in press). Nonetheless, the division has informed much of the scholarship of the wider period, offering an important framework for interrogating the tenth and eleventh centuries (e.g. Richards 2007). That focus has meant that questions remain concerning the nature of the original boundary agreed between Alfred and his Viking adversary Guðrum (McLeod 2014: 231-232). Was this Anglo- Saxon imposition, or a more carefully orchestrated negotiation, incorporating skilled statecraft? This paper aims to elucidate the nature and purpose of the boundary established between Alfred and Guðrum by the agreement commonly known as the Treaty of Wedmore. We suggest that the negotiation and the means by which it was enacted are yet to be properly appreciated. Led by archaeology and history, we embrace the social and political sciences in order to achieve better understanding of borders and their places in landscapes. We begin by introducing the Treaty of Wedmore, its scholarly context, and some of the intellectual problems that surround it. The paper then characterises political contexts, focusing in turn on the strategic positions of Alfred and Guðrum, before discussing the content, semantics, and possible implications of the Treaty. This gives rise to a fuller consideration of the likely geography and topography of the boundary as we understand it, and of its relationship to nearby religious, political, and military sites. Having characterised the boundary arrangements, the paper then explores why Alfred and Guðrum should settle on this rather unusual, even unprecedented, arrangement. From the outset, two questions arise: when was the Treaty agreed, and where? While the Treaty predates Guðrum’s death in AD 890, its date cannot be determined with precision. Hooke (1998: 59) offers 878, following Guðrum’s submission to Alfred after the Battle of Edington. Stenton (1989: 260) suggests it occurred no earlier than 886, while Hadley (2012: 375) opts for c. 880-90. These details have implications: an early settlement might imply an Anglo-Saxon mandate; a later date might reflect a less contested mediation. Nothing in its text confirms that the Treaty was actually concluded at Wedmore. Could it be that arrangements established orally at this site were enshrined in a Treaty of a later date? Or
23 might Wedmore have established a general principle of separation that was only determined by the extent of the subsequent withdrawal of Vikings beyond what became the Danelaw boundary? There is also the intriguing possibility of a return by Guðrum to Wedmore, sometime in the 880s, in order to conclude the Treaty. The Treaty was a binding agreement between two kingdoms separated by culture and ethnicity but sharing the Christian religion. The border (Figure 1) was to be ‘up on Temese, ₇ ðonne up on Ligan, ₇ andlang Ligan oð hire æwylm, ðonne on gerihte to Bedanforda, ðonne up on Usan oð Wætlingastræt’ (University of London/UCL 2019); ‘up the Thames, and then up the Lea to its source, then in a straight line to Bedford, then up the Ouse to Watling Street’ (Whitelock 1979: 416). Implicit are understandings of cartography, border management, harmony of legal structures, and an environment that could facilitate trade. Subsequent tenth- century events notwithstanding, the agreement may have facilitated peaceful co-existence between polities with compatible ideological and administrative worldviews.
Figure 1. The Wedmore boundary.
Understanding of territorial arrangements within the medieval psyche demands discipline in using the terms ‘boundary’, ‘border’ and ‘frontier’; they are not necessarily synonymous. Indeed, some political geographers (Flint and Taylor 2011: 307-309) avoid the descriptor ‘border’, and rather define a boundary as representing the limits of a territory, thereby defining the extent of a state’s sovereignty, and a frontier as a ‘zone at the edge of a historical
24 system where it meets other systems’. Others (Giddens 2013: 50) accept the term ‘border’ but also argue that it is a phenomenon associated only with the development of the nation- state – a construct emerging no earlier than the sixteenth century – presenting the characteristics of a legally determined border, an idea of national identity, and potential for commercial activity with a neighbour (Giddens 2013 and Flint and Taylor 2011, citing Tilly 1975). Geopolitically, the Wedmore concord, with a functioning border, potentially puts Alfred and Guðrum well ahead of their time, as we will discuss later.
A Context for the Treaty The Strategic Picture From the mid-860s, the size of Viking forces arriving in England had increased, and by Alfred’s accession in 871 his kingdom had already experienced years of incursions. Wessex was under threat, but the political geography of Anglo-Saxon England was already complex and in a state of flux. English society had endured warfare between its own kingdoms on a regular basis, with significant political conflict occurring about once every twenty-five years (Lavelle 2010: 15). Northumbria was in turmoil in 866, experiencing civil war when York was lost. Subsequent Scandinavian policy was to rule through a dependent king, Egbert, himself ousted by the Northumbrian Ricsige, who managed to maintain independence for three years until 875/6. Mercia was in decline and was reliant upon Wessex to counter the Viking threat. Alfred had been involved in an inconclusive Mercian campaign at Nottingham in 868 (Swanton 2000, 68-70) before the Vikings took the ancient Mercian seat at Repton, Derbyshire in 873/4. The division of Mercia again saw the installation of a tributary king, the thegn Ceolwulf (Stenton 1989: 252). East Anglia suffered a series of predations, notably the execution of King Edmund in 869. Once more, it seems that governance was subordinated to local kings. Indeed, use of such clients seems to have been a policy applied only to England by ninth-century Scandinavian invaders (McLeod 2014: 173-175). Archaeological corroboration appears in recent coin finds, two from Norfolk attributed to Æthelred of East Anglia, 870-880 (Fitzwilliam Museum 2013). Danish acumen may also be reflected in the appearance of silver pennies from the 890s bearing the legend of the martyred Edmund (Richards 2007: 174). Whatever territorial matrix the kingdoms emerging from the Heptarchy had understood, the model was disintegrating, while Anglo-Saxon and Scandinavian political
25 leaders would have conceptualised the kingdom in markedly different ways. English kings had strategic centres of gravity founded on territorial dominion through royal palaces, while strategic strength for the invaders derived from mobility and a lack of reliance on a specific place. Punitive responses available to Alfred were extremely limited, given that he could not strike at the Scandinavian homelands, even if he had been able to identify them. The concept of formal territorial ownership, within bounded limits, was probably nascent in the Viking strategic psyche, while Alfred was operating within developing legal and geographic paradigms (Stenton 1989: 269-276). Furthermore, there existed across England another relevant model for Alfred (Figure 2): church diocesan boundaries (Reynolds 2002: 68-69), a subject we will revisit below.
Figure 2. Diocesan boundaries of 850 (A Reynolds, 2002).
26 Alfred’s Position Alfred’s kingship was not as secure as is often portrayed. Asser records that in the 850s Wessex was split between the West Saxons and Kent, following a usurpation of Æthelwulf’s authority by his son, Æthelbald. Alfred was the fourth of Æthelwulf’s sons to wear the crown, and might, in other circumstances, have been a regent (his older brother Æthelred, whom he succeeded in 871, already having a son). However, Alfred’s position and subsequent line had been secured in an agreement with Æthelred – documented in Alfred’s will – that, because of the Viking peril, either surviving brother could continue as king (Whitelock 1979: 289- 291, 534). Indeed, Æthelred’s son made two unsuccessful attempts on the throne after Alfred’s death (Stenton 1989: 321-2). A critical period for Alfred began with Guðrum’s arrival at Wareham in 875/6 (Figure 3). Over the next two years Alfred was able to manoeuvre Guðrum to Gloucester, via Exeter, conducting detailed negotiations which involved the unprecedented act of having Guðrum take oaths on a sacred ring (Stenton 1989: 253-254; Swanton 2000: 74). Unaddressed in the Anglo-Saxon Chronicles may be many nuances attached to these mediations, not least in the interpretation of terms such as ‘peace’ or ‘truce’, and for how long they might have been binding (Lavelle 2010: 324-327). Guðrum’s assault on Chippenham in 878 famously forced Alfred to the Somerset marshes and a form of guerrilla warfare (Stenton 1989: 255). The means by which Alfred engineered recovery at Edington/Chippenham are of interest; noteworthy is the manner in which elements of the landscape were used for assembly and mustering points (Lavelle 2010: 5, 179-180). Alfred was developing a sophisticated and perhaps innovative topographical awareness.
Figure 3. Key locations in Wessex and along the Wedmore boundary. Crosses denote religious sites.
27 Alfred needed a solution that offered time and space to counter Viking speed and audacity, while earlier experiences at Nottingham would have added a northern dimension to his threat assessments. We can estimate an army of some 27,000 for Wessex, mustered from the totality of the later burh system of fortified towns developed by Alfred (Lavelle 2010: 60). Hinton (1977: 2) notes the significance of Roman roads and ancient trackways to the strategic picture. At this stage, significant naval deterrence was not part of Alfred’s strategy (Lavelle 2010: 145- 146).
Guðrum’s Position Guðrum appears as part of the Great Army that separated at Repton in 874, going to East Anglia with Oscytel and Anund (Swanton 2000: 72-74). He may have been part of a second wave of invaders not implicated in the earlier martyrdom of Edmund, but Guðrum’s characterisation by the chroniclers remains generally uncomplimentary (Swanton 2000: 72- 76). However, care is needed in handling the grammar of these records, which may have modern connotations inconsistent with Anglo-Saxon sentiment. This quote from Asser is illuminating:
[...] animoseque diu persistens, divino nutu, tandem victoria potitus, paganos maxima caede prostravit, et fugientes usque ad arcem percutiens persecutus est, et omnia, quae extra arcem invenit, homines scilicet et equos et pecora, confestim caedens homines, surripuit, et ante portas paganicae arcis cum omni exercitu suo viriliter castra metatus est. (De Rebus Gestis Alfredi, Asserius).
[…] he defeated with great slaughter and pursued them flying to their fortification. Immediately he slew all the men, and carried off all the booty that he could find without the fortress, which he immediately laid siege to with all his army […] (Lewis-Semple 2006: 16-17).
These words might, in the popular mind-set, describe acts of Viking violence, but they are actually taken from Asser’s heroic description of Alfred’s final subjugation of Guðrum at Chippenham, implying that we do need a balanced appraisal of Guðrum. For example, to what extent was Guðrum implicated in events that provoked Alfred’s maritime campaign to East Anglia in 885 when a major engagement took place at the mouth of the River Stour? The Anglo-Saxon Chronicles are unclear (Swanton 2000: 78). Though these events are clearly post-Edington, they may actually precede the formalisation of the Wedmore agreement. We
28 must remember that Guðrum was but one of many Viking leaders in Northwest Europe, amongst whom a nexus of loyalties would have existed. We know little of the tensions among those groups; it is not credible to assume that they were insignificant. Guðrum may have sought strategic balance through stable borders with Alfred, while then attending to his boundaries to the north and along the coast. We challenge popular dismissals of the post- Edington Guðrum (see, for example, Schama’s characterisation in which Guðrum retires to a life ‘pottering about in un-Viking-like harmlessness’ [Schama 2000: 60-61]). After defeat in 878 and conversion to Christianity, Guðrum became a legitimate king within the Anglo-Saxon framework (Richards 2007: 35). He had lost much of his maritime capability after Wareham (Stenton 1989: 254) but presumably retained a strong base in East Anglia: clearly not one that Alfred, nor Ceolwulf in Mercia, felt willing to challenge. Indeed, Guðrum’s influence over Mercia should not be overlooked; its division into Viking and Anglo-Saxon orbits was only confirmed by Guðrum in Gloucester as late as 877. However, we must question Guðrum’s authority in agreeing boundaries affecting other Viking kingdoms and a supposed client Mercia. Nonetheless, even after defeat, Guðrum retained significant forces, occupying Cirencester for a year after the final reverse at Chippenham (Stenton 1989: 254-257).
Figure 4. Silver penny of Æthelstan II (Guðrum) recovered in Kent, 2006 (Portable Antiquities Scheme database number Kent-338E94, CC-BY-SA 4.0).
We should also address the perpetuation of the name ‘Guðrum’. He was baptised as Æthelstan, with coinage subsequently bearing that name (Figure 4). Yet even in modern scholarship, ‘Guðrum’ is almost universally applied. If his conversion represents such a
29 seminal point, why did the pagan appellation endure? Was Guðrum’s spiritual position ambivalent and were the Chroniclers reinforcing a sense of Viking perfidy in persevering with ‘Guðrum’ rather than the Christianised ‘Æthelstan’?
The Date A Wedmore Treaty, dated 878, would be contemporary with Guðrum’s entry into Christian kingship. His baptism took place three weeks after terms had been agreed. Nevertheless, it was not until the autumn that his forces crossed the twenty-five miles from Chippenham to Cirencester, in Mercia, where they remained for a further year (Stenton 1989: 257). This implies an orderly and dignified withdrawal, on his own terms, incorporating a lengthy stay in territory subsequently to be excluded to Guðrum in the Treaty. A later date of 886 accommodates the problem of a Viking force, possibly there since 872, holding London until then. Independent of Guðrum, that garrison had been established by Hálfdan, a leader of high status. Thus, the eastern flank of the putative English kingdom was still far from secure some years after Edington (Stenton 1989: 246, 258). Until achieving control of London, including the Thames/Lea confluence, Alfred could not have imposed his Wedmore boundary. A more cautious dating of 880-890 (Hadley 2012) potentially leaves us with an even wider gap. Finally, we should address the relevance to Alfred and Guðrum of a definitive date. According to the surviving text, the Treaty made between the two kings and their people was ge forgeborne ge for ungeborene (University of London/UCL 2019) ‘both for the living and those yet unborn’ (Whitelock 1979: 416). Perhaps agreed over a number of years, there was no purpose in recording a date, while some ambivalence may have been politically expedient. Hadley (2002: 16-17) notes that such treaties were beyond the experience of illiterate, yet adaptable, Danish societies. However, illiteracy need not bar sophistication in the arts of soft as well as hard power.
The Content of the Treaty This concord offered arrangements for peaceful coexistence, establishing confidence-building measures and opportunities for future contact and, possibly, trade. While the historical picture presents descent to violence after Guðrum’s death, this may have been contrary to the aspirations of the two protagonists, and the use of text may have been as attractive to Guðrum as it was to Alfred (McLeod 2014: 210-213). The two surviving versions of the Wedmore
30 Treaty (Thorpe 1840: 66-67) notably contain variation in text. Though geographical points of reference are consistent, the precise wording raises some ambiguity. Scholarship generally relies on Whitelock (1979: 416-417), her translation being slightly more in the modern idiom. This extract from Thorpe (1840: 66) develops the point:
Ærest ymb ure landgemæra : up on Temese, ₇ ðonne up on Ligan, ₇ andlang Ligan oð hire æwylm, ðonne on gerihte to Bedanforda, ðonne up on Usan oð Wætlingastræt (University of London/UCL 2019).
First, concerning our land-boundaries: up on the Thames, and then up on the Lea, and along the Lea to its source, then right to Bedford, then up on the Ouse unto Watling-street.
Thorpe’s version specifically refers to ‘ure landgemæra’ (‘our land-boundaries’). The semantics may be useful in better comprehending the relationship between the two kingdoms. The threat to Wessex had been maritime as well as land-based or riverine. It may be that Wessex’s implicit sovereignty of the south coast and Kent was sufficient for Alfred. However, we are left with a difficulty in determining at what geographic point the Thames would have become an effective means of delineating a boundary. The estuary is almost 20 kilometres wide between Shoebury Ness in Essex and the Medway in Kent, narrowing to c.300 metres at the Lea’s confluence. Passage along the Thames would probably have been essential to both parties. Did Alfred believe that Guðrum posed no threat to Kent and sites such as Canterbury, or was there a general assumption that the coasts of the two kingdoms of Wessex and East Anglia presented de facto, if not de jure, borders? The Thames was certainly a logical starting point for both sides and, importantly when establishing international/inter-kingdom concords, a point of agreement around which the rest of a negotiation might cohere. Thorpe’s translation also uses the terms ‘up on’ when referring to all three rivers and, in addition, uses ‘along the Lea to its source’. His translation of the second version (Thorpe 1840: 66) says the boundary runs: Along the Thames, then up on the Lea, along the Lea unto its source, thence right to Bedford, thence upwards on the Ouse unto Watling-street. These variations offer clues as to how the boundaries were to be respected. A river is more than a division between two pieces of land. It may be rich in resources as well as being an important line of communication, while extended or shifting floodplains or banks may introduce confusion. Modern maps will specify, for example, which bank marks the border and whether it is at the high or low water mark. The border may alternatively follow the thalweg of the
31 river (the deepest point of flow along the riverbed, and not necessarily coincidental with the point of mid-stream.) Where a river defines a boundary, we should consider issues such as access or the right to draw water. One interpretation available from the words ‘up on’ is that they meant ‘as far as’ the respective home banks of the river, rather than the course of the river itself. Thus, each party would be able to approach up on the banks of the Lea and along the banks to its source. ‘Along the Thames’ could suggest use of that river for both parties. Tides, sand banks and currents may have necessitated local concessions towards the other’s shipping. The width of the estuary and relative lack of settlements may have made this a tolerable compromise. However, along the Lea and Ouse, the two sides were in much closer proximity and we should consider the possibility that waterways could represent neutral, if not forbidden, zones. The use of rivers is itself significant. Citing Dalton (2006), Lavelle (2010: 320-321) emphasises the importance of rivers in conducting negotiations as well as their potent symbolism as insuperable obstacles. The concept of a riparian boundary would have resonated across Christian, pagan and militaristic ideologies. Notably, wetlands and rivers offered an important medium for ritual deposition, especially of weapons, in southern Scandinavian culture (Pedersen 2012: 208-209). We also need to consider Whitelock’s ‘straight line to Bedford’ running from the source of the Lea. This twenty-five-kilometre lineal division is unprecedented in determining national frontiers in the British Isles, being characteristic of some boundaries established within post-medieval and colonial geo-politics. Thorpe differs in translation, using the phrase ‘thence right to Bedford’. This might imply a boundary following the most direct and easily recognisable route, accommodating local deviations. Also, as the line runs almost due north, the absence of the cardinal point is at odds with its common use in describing direction in the late ninth century (Bately 2007: 44-47). The Treaty’s clauses are equally important research assets. They established an equivalence of the law codes in the two kingdoms, both in terms of levels of recompense and how transgressions should be investigated. Furthermore, they explicitly set conditions under which contact and even trade could be conducted (Whitelock 1979: 416-417). A desire for peaceful co-existence seems implicit.
32 A Landscape Perspective A topographical and archaeological review of the boundary, particularly that section running from Leamouth to the Ouse/Watling Street junction, is essential in divining its purpose. Here we proceed in stages along the boundary and, while scrutiny of Watling Street - potentially as far as Chester - is too great an undertaking for this study, some comments on Alfred’s position regarding his relationship with Mercia and the more northerly Scandinavians are offered.
The Lower Lea The Lea enters the Thames in London’s East End, well downstream of the settlement of London captured by Alfred in 886. Importantly, those expelled Vikings are likely to have been from a contingent not associated with Guðrum (Stenton 1989: 258-259). The Lower Lea possibly defined the ancient Essex/Middlesex border, but any historical evidence derives from the much later Domesday survey (Campbell 1962: 97). Archaeology in the heavily urbanised Lower Lea Valley is sparse, but a number of observations are possible. Passage was possible upstream for 20 miles. A forty-foot clinker built boat, found in 1900 during work on Lockwood Reservoir, Walthamstow (English Heritage Pastscape 405588), may have been part of the Viking fleet that sailed upriver in 896 (Whitelock 1979: 205), but the river was probably not extensively navigable until twelfth-century improvements made by the Abbot of Waltham (Baxter 2011a: 55). The possibility of a land corridor along the valley is supported by the recovery of a Viking stirrup near the east bank at Stratford in 1864 (English Heritage Pastscape 962891). Domesday accounts suggesting settlement and exploitation of the Lower Lea (Darby 1952: 228-246) contrast with archaeological results from the site of London’s Olympic Park, 2005-09. Excavations yielded disappointing results (Pitts 2012: 16) – particularly for the early medieval period – though there are indications of limited agriculture, a fast-flowing river and intermittent flooding (Archer and Spurr 2009: 14 and 19). Today, much of the lower valley is inundated where perhaps significant archaeology is preserved. Astride the Lea, the landscape is potentially informative, especially where the river itself closed for access (Figure 5). Adjacent to the west bank is Ermine Street, another potential boundary candidate. This afforded a valuable north-south axis from London to Hertford, while offering no penetration into Wessex. On Guðrum’s side a route parallel to the Lea was offered by the River Roding. Between the Roding and Lea is a prominent ridge which would have dominated both rivers
33 (Baxter 2011b: 62) with Iron Age and Roman features at strategic points; Loughton Camp gave oversight of the Thames and Kent. Uphall, on the Roding, may even have been altered by the Danes (Chalkley Gould 1903: 277-282).
Figure 5. The Lower River Lea.
The Upper Lea Little suggests the Upper Lea was an ancient boundary. Essex was probably defined by the River Stort beyond its confluence with the Lea (Darby 1952: 211), while Middlesex may have had a northern border running through a heavily wooded area, separating it from what is now Hertfordshire (Campbell 1962: 97). However, for Alfred and Guðrum, the Upper Lea presented an unambiguous feature, straddling any potential east-west axes. As an area of considerable importance to the Church (Figure 3), the Upper Lea included Hertford, where Bede records the Synod of 673. Here Theodore, Archbishop of Canterbury, convened an assembly of bishops from all the English kingdoms, confirmed the organisation of the Church, and set the date of Easter (McClure and Collins 2008: 180-183). Notably, the Synod was establishing a consciousness of ‘Englishness’ which would not have been lost on Alfred,
34 while diocesan boundaries were being established that were not coterminous with either the Anglo-Saxon kingdoms or the later Danelaw (Figure 2). Also, in Theodore’s document, a doctrinal dispute is formally resolved, followed by the imposition of mechanisms designed to prevent tensions between religious houses and bishoprics. This presents striking resonance with similar arrangements laid out in the Treaty. Upstream lies Hatfield, site of the Council of 680, which confirmed the orthodoxy of the Church in England (McClure and Collins 2008: 199-200) and was important in both continental and English contexts. Thus, we find eminent religious centres included along a border defining two Christian kingdoms, perhaps functioning as both spiritual and physical portals. Given the Christian context, we might look to episcopal structures as providing one means by which Wedmore could work. This is put into stronger relief by considering Hatfield’s neighbour, the monastery of St Albans, on the Wessex side of the Lea. The cult of St Alban was central to Christianity in Britain, the site importantly attesting continuity of religious worship (Crick 2007: 3-7). Knowledge of St Albans’ landholdings would be helpful, but this area and period is poorly covered by surviving Anglo-Saxon charters (Reynolds 2002: 29-32). Mercian kings made grants to St Albans which may have included estates in Hertfordshire beyond the Lea (Sawyer 1968: 105-119). The scant and later records of the Charters of St Albans comprise much of the documented evidence. From these we can infer significant grants from Offa, while Domesday Book shows the abbey to have held estates on either side of the river (Crick 2007: 35-76).
The Source of the Lea For some distance below its source, the Lea presented little obstacle to movement. The source lies at the northern end of Waulud’s Bank (Figure 6), a large Neolithic enclosure (Goddard 1904: 267-268). An important point of reference in an area that had featured in military campaigns since the sixth century (Swanton 2000: 18-19), it is preserved in the Limbury suburb of Luton. From here we can take a wider topographical view of the border, incorporating higher ground on the Dunstable Downs to the west, and the Warden Hill/Galley Hill ridge to the east. The panorama gained from Five Knolls on the Dunstable Downs is informative. This escarpment would have dominated Waulud’s Bank, the border towards Bedford and the lowlands accessing the upper Thames Valley deep to the west. It gave strategic oversight of much of Watling Street which, like Ermine Street, gave Alfred a vital lateral route that would
35 have offered nothing to any westward advancing force. The Downs overlook the Ickneild Way and present an ideal beacon site from which warning could be sent deep into Wessex, via Ivinghoe Beacon clearly visible to the west. Although archaeological evidence is limited, the use of beacons by the Anglo-Saxons is almost certain (Rumble and Mills 1997; Lavelle 2010: 218-225). Adjacent to Watling Street are the Roman enclosure at Totternhoe and the Neolithic/Roman earthworks at Maiden Bower. To the east, Guðrum similarly controlled higher ground. From the Warden Hill/Galley Hill ridge he had oversight of the Lea and the boundary towards Bedford. He also had possession of the Ickneild Way as it passed into East Anglia. Thus, a key point of defensive parity seems to have been achieved.
Figure 6. The Upper Lea area and the line to Bedford.
Five Knolls and Galley Hill are intervisible and at a very similar elevation (205 metres and 185 metres Ordnance Datum [OD]), a bearing between the two running directly through Waulud’s Bank. The relationship between these locations is reinforced given that both are the site of (probably pre-Alfredian) Anglo-Saxon execution cemeteries with interments cut
36 into Bronze-Age round barrows (Reynolds 2009: 97-103). Contemporary Scandinavian cremation mounds at Heath Wood, Derbyshire, and inhumations at the captured Mercian royal centre of Repton (Richards 2007: 195-201) both imply Viking statements of geographic and cultural control. It is tempting, therefore, to interpret the Luton burial sites as having some role in the Wedmore negotiation.
The Lea Source to Bedford The boundary fails to respect any topographical features over its twenty-five kilometres north from Waulud’s Bank where, at 120 metres OD, the ground rises gently for about four kilometres. From here, the Chiltern escarpment drops, the land undulating to thirty metres OD at Bedford. The line passes close to the Iron-Age fort at Sharpenhoe Clappers but otherwise sites of significance seem to exist only further to the east. English linear boundaries before and during the Early Middle Ages share certain characteristics. If of a large scale, they tend to involve earthworks or other adjustments to the landscape – such as stone avenues – and are rarely geometrically straight (Muir 2002: 67- 75). Hooke (1998: 160-1 and 215-7) notes the development of lines of communication comparable in length to the Leagrave-Bedford line associated with droving. They are not ‘straight’, but this may not be what was intended by the phrase used in the Treaty. Some important religious routes were probably designated with stones, or by reference to points such as holy trees (Blair 2006: 480-481). There is nothing immediately suggestive of straight lines in the management of the landscape, but closer inspection may hint at boundary delineation. Development to the north of Waulud’s Bank appears to respect an old trackway, partly marked by an avenue of trees, and clearly visible in 1940s aerial photography (Figure 7). The orientation is towards Bedford. Terminating after about a mile, it appears in modern Ordnance Survey and mapping of 1834. It is possible that the track was associated with the Neolithic enclosure, and marks the start of an ancient route. It may also perpetuate a ninth- century attempt at boundary definition. Similar hints from hedgerows and rights of way also extend towards Bedford from Sharpenhoe Clappers. Finally, it is worth noting an observation by Gore Chambers (1904, 18) of a much earlier ecclesiastical division of Bedfordshire. This arrangement had seen the south of the county under the control of the West Saxon bishopric of Dorchester, while the remainder
37 passed to the diocese of Leicester. Archaeologically, this remains unresolved but cannot be dismissed as a factor in determining our border.
Figure 6. Waulud's Bank, bottom centre, with treeline and path/hedgerow extending towards Bedford (Historic England).
Bedford The use of Bedford as the terminus for this line is intriguing. Relatively isolated in Roman times, with only limited pottery or tile manufacturing and some mineral extraction (Ordnance Survey 2001), it had attained both secular and ecclesiastical importance by the ninth century. The local church seems to have been attaining minster status, Bedford is named as having elders of some importance, and Offa is reputedly buried there (Blair 2006: 130, 318 and 331). Excavations and stonework from Bedford’s churches suggest Saxon origins, and it may have served as an emporium (Edgeworth 2007: 95-96). Its status may be advanced by recent work incorporating the River Ouse, associated water-flows and local defensive earthworks (Edgeworth 2011: 24-26). This confirms the location of the early burh on the northern bank, with the southern fortifications appearing only after Edward the Elder’s campaign of 914-15.
38 In Bedford, Alfred relinquished a site of considerable value. The alternative is to suggest an Alfredian salient on Guðrum’s side of the river which becomes difficult to reconcile with the Chronicles’ narrative of Edward the Elder retaking the town.
The Ouse to Watling Street Above Bedford the Ouse is shallow and meandering. While a direct line from Bedford to Watling Street is less than thirty kilometres, the tortuous route along the river is about eighty kilometres in length, and navigable only by short boats of shallow draught. This sector remains almost devoid of archaeological evidence, with the exception of Harrold, on the north bank. Marked by eighteen ring ditches from the Bronze Age, there is evidence of Anglo-Saxon occupation in the seventh and eighth centuries, as well as a probable ninth- century furnished Viking burial, incorporating a sword, bead, iron bucket and buckle (English Heritage Pastscape 346787).
Watling Street to Chester At the close of the ninth century, Alfred’s daughter Æthelflæd controlled Mercia as far north as Chester and was initially prepared to deal peaceably with Vikings settling in the Wirral. Archaeology supports the identification of Cheshire as a frontier area (Griffiths 2007, 402- 405), but there is no evidence to confirm that the border extended this far. Indeed, the terminus for Watling Street was probably Wroxeter (Ordnance Survey 2001). Given the limits of Guðrum’s kingdom, perhaps ambiguity was intended. Nor should we discount the possibility of ambitions for a greater Wessex defined in the west by Offa’s Dyke.
The Burghal Hidage and Mercia The Danish threat stimulated the development of defended towns or burhs across Wessex. At least thirty-three sites are known from the Burghal Hidage (c 914-18), no village being further than thirty-two kilometres from support (Stenton 1989: 264-265). The Hidage significantly post-dates Wedmore, but the burhs deserve mention for a number of reasons. Burhs are likely to have been an Alfredian innovation even if completed by Edward the Elder, with consensus for their dating insecure (Hall 2011: 606). Davis (1982: 807) addresses problems in the Hidage dating for the burhs, suggesting that later Mercian inclusions, such as Buckingham, may actually have been incorporated in Alfred’s plans as early as 886/90. Notably, the Anglo-Saxon Chronicles fail to record Buckingham’s apparent loss to the Danes
39 relatively soon after the Treaty would have been concluded. The Chronicles do, however, go on to record that this town – Anglo-Saxon according to the Treaty – was recovered from the Danes in 912-14 (Davis 1982: 804-807). Davis’s thesis may not be widely accepted, but nonetheless presents a context for the boundary remaining unclear regarding Watling Street. Alfred’s main focus was on Wessex, and unequivocal protection of Mercia may have been beyond his resources, leaving Watling Street beyond the Ouse as a frontier area. As a logic for Alfred begins to emerge, we now address territorial strategies that may have been available to Guðrum. To supplement documentary evidence, here we draw on evidence from Scandinavian archaeology and make informed analogy from ethnography. Such study of non-Western societies is a well-tested means of exploring diverse behaviours and, while it has limitations, it allows us to escape the orthodoxies of twenty-first-century European society.
Alternative Viking Options Assuming Guðrum influenced the negotiation, we need to consider what alternatives he might have had in defining his kingdom. Doing so challenges some assumptions regarding what the archaeological record says about territorial control. Comparanda are essential in comprehending the evolving mentalities in early medieval state development. Here, we discuss broadly contemporaneous alternative means of demarcating territory, including examination of sites likely to have been occupied by Guðrum, and conclude by offering Wedmore as a model for future negotiations.
The Danevirke The multi-phase Danevirke is a wall of up to three metres in thickness, probably dating from the seventh century (Schultz 2010). In the late ninth century it extended westwards along the base of the Jutland peninsula from the then unfortified Baltic emporium of Hedeby (Hilberg 2012: 106). Some thirty kilometres long, it is similar in length to the Leagrave- Bedford line. An obstacle to movement, the Danevirke could not precisely delineate a border. Its construction, maintenance and the need to control approaches to it would place it within home territory. Functioning as a symbol of territoriality and control rather than a specific boundary, even approaching it may have been deemed a hostile act. However, it is probably best understood as a means of facilitating a safe line of commerce between the North and Baltic Seas (UNESCO 2011). During the ninth century a function of strategic defence may
40 be questioned. Ships and horses meant that dykes could be easily bypassed. In England, the Wansdyke had become militarily irrelevant, while the Cambridge Dykes, important post- Roman military defences (Storr 2013: 48-49), appear only as a geographical point of reference. Chroniclers describe Edward’s campaign of 903, in which ‘he harried all their land between the Dykes and the Ouse’ (Whitelock 1979: 208).
The Longphuirt Riverine longphuirt sites in Ireland may aid our understanding of Viking territoriality. These sheltered locations (often incorporating a tributary for enhanced defence), gave access to the sea while providing protection for longboats and their crews. Sites at Woodstown, Dublin and Dunrally show them as typically up to 360 metres by 150 metres, forming a D-shaped enclosure, with outer ditches of five metres in depth and up to two metres wide (Wallace 2012: 435-437). Ireland was more loosely governed by smaller, kin-based kingdoms, none of which were comparable in scale to Northumbria or East Anglia, who capitulated after short campaigns. The Church in Ireland was also much more aggressive in its response to Viking raids (Ó Corráin 2012: 429-430). Against this background, evaluation of the longphuirt is difficult, especially given that sites so far discovered may only represent a fraction of the total constructed. The possibilities for longphuirt could range from temporary outposts to inter- connected sites controlled at polity level, and whose purpose was to both define and declare territorial ownership. The longphuirt may help in assessing riverine sites near the Leagrave-Bedford line (Figure 6), generally reported in the context of early tenth-century Anglo-Saxon campaigns (Goddard 1904: 280-284; Stenton 1989: 327-328). Places and times may have become confused, an example being an Anglo-Saxon Chronicle reference to Viking fortification of Tempsford at the Ouse/Ivel confluence in 917 (Richards 2007: 38). Tempsford has archaeological evidence of settlement planning that is contemporary with this period, but nothing of precise Viking provenance (Maull and Chapman 2005: 5, 21), although pending archaeology may prove positive (Edgeworth and Raffield 2008). Another candidate for the 917 fortification is Beeston (Richards 2007: 39), five kilometres upstream on the Ivel, and much closer to the river’s navigable limits. However, the rivers and water table may have changed significantly since the early Middle Ages. Richards (2007: 38-39) and Goddard (1904: 276-280) refer to a number of sites retaining only a trace of a waterway. One is Shillington/Church Panel, described by Goddard
41 as ‘an island in the midst of a former swamp’ (Goddard 1904: 276-280). It presents a highly visible D-shaped enclosure. Most significant is arguably Willington. Probably the final point on the Ouse at which large boats could be conveniently manoeuvred, it may have provided docking for up to thirty ships. This site, the largest of its type in the region, is not mentioned in the Anglo-Saxon Chronicle, suggesting that Willington was simply abandoned when Bedford was lost (Goddard 1904, 284). There are other interpretations for the site, which has produced quantities of St Neot’s ware but nothing definitively Scandinavian. Unfortunately, much has been destroyed by rail workings and modern horticulture (English Heritage Pastscape 362989). The proximity of these sites to the border is noteworthy. Plausibly operating in a similar manner to the longphuirt by the time of Edington, they may subsequently have played a role in setting the Leagrave-Bedford line. They also represent a rather limited military investment by Guðrum, who may have included ancient places such as the Iron-Age fortifications of Ravensburgh and Sharpenhoe Clappers in his defensive calculations. If Guðrum’s strategic reckonings were not influenced entirely by Wessex, some resources may also have been directed towards his lands neighbouring other Viking entities beyond the Ouse and Watling Street.
D-Shaped Enclosures Excepting Shillington (whose Viking attribution is only circumstantial), none of Guðrum’s sites displays a convincing D-shaped enclosure, though some provisional conclusions suggest Tempsford as a possibility (Edgeworth and Raffield 2008: 1). There are several possible explanations for this absence. Perhaps the occupants did not feel sufficiently threatened to invest in the work? Perhaps they were defensible without significant earthworks? It may also be that the D-shaped enclosure is a template too readily applied to England’s Viking sites, and the form may not be as common as is generally supposed. Viking forces can leave scant archaeological signatures, even in terms of physical defences (Hadley and Richards 2016). This may be the case at the Great Army’s winter camp on the Trent at Torksey, Lincolnshire. Historical and artefactual evidence is plentiful, yet there is nothing to suggest investment in significant man-made defences (Figure 8). Torksey may prompt a review of other sites, including Repton.
42
Figure 7. Geophysical survey of the Great Army site at Torksey (H Brown).
Repton is a key site in understanding Viking operations in England. The existence of earthworks here is beyond doubt, with construction sometime between the Middle and Late Anglo-Saxon periods (Biddle and Kjølbye-Biddle 1992: 40). However, the extrapolation of limited geophysical survey data into a neat enclosure is perhaps optimistic. Repton provides two difficulties. First, can we confidently assert the enclosure as essentially Viking work? Excavation shows it to have had four successive ditches (Biddle and Kjølbye-Biddle 1992: 40). The earliest may have been Mercian with some earthwork defences created to counter the Viking threat. Second, barely measuring one hundred metres by sixty metres — and considerably smaller than the longphuirt — the structure is unconvincing as a major bastion and could only have accommodated a limited component of the Great Army (Hadley and Richards: 2016). Here it is also important to acknowledge the reappraisal of the dating of burials at Repton by Jarman et al (2018) plausibly bringing a greater number to a single event or occupation contemporary with the Great Army.
43 Non-Western Boundaries In non-Western societies boundaries can be specifically delineated as well as zonal. Examples come from regions as diverse as Alaska and New Guinea, where trespass can be fatal (Diamond 2012: 42). Effective boundaries are contingent upon certain conditions. The population-to-border ratio must be sufficient to allow boundary patrolling or oversight without excessive cost. Group membership or nationality should be clearly defined, even if cross-border social contact is permitted. Each territory should provide sufficient resources to support its population without recourse to raiding a neighbour, and a defined territory is likely to contain something worth defending (Diamond 2012: 37-43). Study of such societies may help to inform the Viking experience across the British Isles. While it remains difficult to identify precise intentions regarding territorial definition, it is possible that they were influenced by knowledge of earlier west-Scandinavian contact with the Sámi people. Native to central and northern Norway and Sweden, the Sámi provide important clues as to how Vikings mediated cultural boundaries (Price 2017) and it is credible to view these two separate societies existing in a state of symbiosis. The Sámi provided important goods such as pelts, boats and skis and are unlikely to have been considered entirely as a source of plunder (Olsen 2003: 10 -27). This ambivalent relationship, which saw cultural tolerance and intermarriage, was reflected by borders that lacked precision but were mutually understood. Marked by Sámi burials and hoards, boundaries seem to have been symbolic as much as geographic (Zachrisson 2012: 32-37) and indicative of zonal frontiers. The Sámi/Norse relationship was known to Alfred through contact with the Norse traveller Ohthere, probably received at Alfred’s court sometime after Edington (Sawyer 2007: 139). Ohthere displays extensive understanding of European geography and the use of rivers, such as the Vistula, as boundaries between kingdoms. Moreover, Viking expansion and settlement variably involved conflict, trade and tribute (Bately 2007: 46-48), and Ohthere’s report demonstrates Alfred’s preparedness to engage with ambassadors of potentially hostile kingdoms. Indeed, Alfred’s emissaries may even have been accepted at York, deep into the Danelaw (Hadley 2000: 10-11). It is highly plausible, therefore, to see Alfred as imbued with a geopolitical and diplomatic perspicacity that had been informed by world views beyond the normal ambit of Anglo-Saxon Christianity.
44 Another Border? Subsequent events in early tenth-century Normandy are noteworthy, rehearsing termination of conflict, the establishment of borders and the process of Viking conversion. Between 911 and 933, Scandinavian settlement was legitimated through formal grants of land by the West Franks, most notably in the Treaty of Saint-Clair-sur-Epte. The Treaty’s date is unclear, and in the absence of surviving documents we rely on the later records of Dudo of St Quentin (Renaud 2012: 453-454). Again, we may have boundaries only partially conforming to geographic features. The protagonists, the Viking Rollo and King Charles the Simple, probably met in a formal manner, but this time Rollo was militarily ascendant. The likelihood is that he, or an associate, had contact with Guðrum or Alfred, having sailed to England where, according to Dudo, ‘he entered into relations with ‘Alstelmus’, the most Christian King’ (Douglas 1942: 418, 426-429).
A Move Towards Geo-Political Innovation Finally, the Wedmore arrangement begs comparison with British earthworks associated with borders. Particularly informative is Offa’s Dyke, a multi-phase construction that is certainly not archaeologically ephemeral. The Dyke as a frontier would have resonated in contemporary kingship. It was Alfred’s friend Asser who recorded that ‘Offa ordered a great vallum to be made from sea to sea between Wales and Mercia’ (Stenton 1989: 213). However, while the Dyke can be characterised as a defensible border, variously respecting or ignoring physical features, determining exactly what constitutes Offa’s Dyke is problematic. To the northeast lies Wat’s Dyke, possibly of the earlier eighth century (Hill and Worthington 2003: 38-45), although charcoal radiocarbon dating suggests even earlier construction (Feryok 2011: 165). Recent work at Chirt may even put some of the main dyke well before Offa (Belford 2014). It is Wat’s Dyke, not Offa’s, that reaches the sea. Hill and Worthington (2003: 47, 108-111) also argue for smaller features from the Herefordshire Plain to the Severn Estuary as being of the design of someone other than Offa, with only the works from Rushock Hill (Herefordshire) to Treuddyn (south of Mold) constituting the Dyke proper. While the earthworks may not have stretched from sea to sea, the River Wye may have been incorporated within the frontier or border. The Ordinances of Dunsaete (c. 926) – an intriguing document agreed between the English Witan and Welsh counsellors – sets conditions by which a river boundary, probably the Wye, was to be managed (Feryok 2011: 172-173). The Ordinances bear remarkable textual similarity to the Treaty of Wedmore.
45 These Western dykes operated within political as much as military domains in which Offa viewed himself as ‘Rex Anglorum’ (Stenton 1989: 212). Asser’s idea of the ‘sea to sea’ boundary is also highly relevant, being similar in scale to the Danelaw divide. It would be a small step to connect the northern extremities of the Wedmore and Offa designs and tempting to consider such a contiguous border as defining an Alfredian English kingdom, with Asser as an agent in that design. Undertakings like Offa’s Dyke were the work of years, probably coinciding with periods of peace (Feryok 2011: 185). The concept of a dyke may have been attractive to Alfred but the terrain bordering Guðrum’s kingdom was often low and flat, with higher ground generally biasing north-south oversight over the essential east-west. Manpower could have been provided through the Hidage structure (Hill and Worthington 2003: 116-126) but at the cost of an effective burh scheme. Time and resources militated against major construction. Political sophistication could offer novel and effective solutions.
Conclusion The Treaty of Wedmore suggests adroit mediation providing unambiguous delineation between two Christian polities within a legal framework, an important element of which may have been conflict prevention and the peaceful resolution of disputes. This was a border agreed between two sovereign kings rather than a zonal frontier. It existed within a milieu of kingdoms, cultures in conflict and various stratagems for territorial negotiation. It also sits within an important European dimension, involving the dynamics of nascent state formation and inter-state contact. Little archaeology has been conducted which explores the collaborative possibilities for Alfred and Guðrum. Instead, the paradigm of the Wessex burhs has dominated along with the subsequent campaigns of Edward the Elder. We need a reconsideration of Guðrum/Æthelstan’s position within contemporary Viking kingship and a more informed view of Alfred’s strategic intent. A key issue remains the degree to which we believe that Guðrum’s kingdom was truly Christianised, at least politically. Important archaeological agendas arise in explicating the nature of Guðrum’s East Anglia. Not least would be the collection and interrogation of data from monuments, buildings and settlements (religious and secular), from which cultural signatures and power bases might be inferred. As an early example of sophisticated arbitration, the Alfred-Guðrum division demands careful consideration that incorporates not only a legal/textual analysis, but
46 attention to landscape and strategy. An understanding of how those borders were negotiated may help to elucidate the dynamics behind the Wedmore settlement. Moreover, if we can better comprehend the mechanics of border control, and the associated material signatures, we may be able to develop a more nuanced approach to archaeological investigation of these landscapes.
47 Notes
1 In this paper, ‘Viking’ refers broadly to Scandinavian-led groups campaigning into North-West Europe. For close discussion of the complexities of that descriptor see Jesch (2015).
48 Bibliography
Asserius. 2012. De Rebus Gestis Alfredi (Scotts Valley: Create Space Independent Publishing Platform) Baker, John, and Stuart Brookes. 2012. ‘Fulham 878-9: A New Consideration of Viking Manoeuvres’, Viking and Medieval Scandinavia, 8: 23-52 Bately, Janet. 2007. ‘Text and Translation: The Three Parts of the Known World and the Geography of Europe North of the Danube According to Orosius’ Historiae and its Old English version’, in Ohthere’s Voyages: A late 9th-Century Account of Voyages Along the Coasts of Norway and Denmark and its Cultural Context, ed. by Janet Bately and Anton Englert (Roskilde: The Viking Ship Museum), pp. 40-58 Baxter, Alan. 2011. ‘Lea River Valley’ in London’s Natural Signatures: The London Landscape Framework, ed. by Alan Baxter and Sheils Flynn (London: Natural England), pp. 54-57 Baxter, Alan. 2011. ‘Roding River Valley’ in London’s Natural Signatures: The London Landscape Framework, ed. by Alan Baxter and Sheils Flynn (London: Natural England) pp. 62-65 Bede. 2008. The Ecclesiastical History of the English People, trans. by Bertram Colgrave and ed. by Judith McClure and Roger Collins (Oxford: Oxford University Press) Biddle, Martin and Birthe Kjølbye-Biddle. 1992. ‘Repton and the Vikings’, Antiquity, 66 (25): 36-51 Blair, John. 2006. The Church in Anglo-Saxon Society (Oxford: Oxford University Press) Brink, Stefan. 2012. ‘Law and Society. Polities and Legal Customs in Viking Scandinavia’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 23-31 Brookes, Stuart and Andrew Reynolds. 2019. ‘Territoriality and Social Stratification: The Relationship between Neighbourhood and Polity in Anglo-Saxon England’, in Polity and Neighbourhood in Early Medieval Europe, ed. by Julio Escalona, Orri Vésteinsson and Stuart Brookes (Turnhout: Brepols), pp. 267-304 Campbell, Ella Muriel Joice. 1962. ‘Middlesex’, in The Domesday Geography of South-East England, ed. by Henry Clifford Darby and Ella Muriel Joice Campbell (London: Cambridge University Press), pp. 97-137 Chalkley Gould, Isaac. 1903. ‘Ancient Earthworks’ in The Victoria History of the Counties of England: A History of Essex, Volume I, ed. by Arthur Doubleday and William Page (Westminster: Archibald Constable), pp. 275-314 Crick, Julia. 2007. Anglo-Saxon Charters XII: Charters of St Albans (Oxford: Oxford University Press) Dalton, Paul. 2006. ‘Sites and Occasions of Peacemaking in England and Normandy 900-1150’, Haskins Society Journal, 16: 12-26 Darby, Henry Clifford. 1952. The Domesday Geography of Eastern England (London: Cambridge University Press) Davis, Ralph H. C. 19082. ‘Alfred and Guthrum’s Frontier’, The English Historical Review, 97 (385): 803-810 Diamond, Jared. 2012. The World Until Yesterday: What Can We Learn From Traditional Societies? (London: Penguin)
49 Douglas, David Charles. 1942. ‘Rollo of Normandy’, The English Historical Review, 57 (208): 417-436 Edgeworth, Matt. 2007. ‘‘Anglo-Saxon’ and Medieval Bedfordshire’, in Bedfordshire Archaeology: Research and Archaeology: Resource Assessment, Research Agenda and Strategy, ed. by Martin Oake, Mike Luke, Michael Dawson, Matthew Edgeworth and Peter Murphy (Bedford: The Bedfordshire Archaeological Council and Bedfordshire County Council), pp. 87-117 Edgeworth, Matt. 2008. ‘The Tempsford Project: An Interim Report’, South Midlands Archaeology, 38: 8-16 Edgeworth, Matt. 2011. ‘The Weir and the Flowing Earthworks of Bedford’, British Archaeology, 121: 22-27 Feryok, Marge. 2011. ‘Offa’s Dyke’ in Mercia: The Anglo-Saxon Kingdom of Central England, ed. by Sarah Zaluckyj (Almeley: Logaston Press), pp. 163-192 Flint, Colin and Peter Taylor. 2011. Political Geography: World Economy, Nation-State and Locality (Harlow: Pearson Education) Giddens, Anthony. 2013. The Nation-State and Violence: Volume Two of a Contemporary Critique of Historical Materialism (Hobeken: Wiley) Goddard, A. R. 1904. ‘Ancient Earthworks’, in The Victoria History of the Counties of England: Bedfordshire, vol. 1, ed. by William Page (Westminster: Archibald Constable), pp. 267-304 Gore Chambers, C. 1904. ‘Political History’, The Victoria History of the Counties of England: Bedfordshire, vol. 2, ed. by William Page (Westminster: Archibald Constable), pp. 17-72 Griffiths, David. 2011. ‘The Early Medieval Period’, in Meols: The Archaeology of the North Wirral Coast. Discoveries and Observations in the 19th and 20th Centuries, ed. by David Griffiths, Robert Philpott and Geoff Egan (Oxford: Oxford University School of Archaeology), pp. 399-406 Hadley, Dawn M. 2000. The Northern Danelaw: Its Social Structure, c.800-1100 (London: Leicester University Press) Hadley, Dawn M. 2002. ‘Invisible Vikings’, British Archaeology, 64: 16-21 Hadley, Dawn M. 2012. ‘The Creation of the Danelaw’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 375-378 Hadley, Dawn M. and Julian D. Richards. 2016. ‘The Winter Camp of the Viking Great Army, AD 872-3, Torksey, Lincolnshire’, The Antiquaries Journal, 96: 23-67 Hall, Richard. 2011. ‘Burhs and Boroughs: Defended Places, Trade and Towns. Plans, Defences, and Civic Features’, in The Oxford Handbook of Anglo-Saxon Archaeology, ed. by David A. Hinton, Sally Crawford and Helena Hamerow (Oxford: Oxford University Press), pp. 600-621 Haslam, Jeremy. 2005. ‘King Alfred and the Vikings: Strategies and Tactics, AD 876-886’, Anglo- Saxon Studies in Archaeology and History, 13: 122-54 Haslam, Jeremy. 2012. Urban-Rural Connections in Domesday Book and Late Anglo-Saxon Royal Administration: BAR BS, 571 (Oxford: Archaeopress) Hilberg, Volker. 2012. ‘Hedeby: An Outline of its Research History’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 101-111
50 Hill, David and Margaret Worthington. 2003. Offa’s Dyke: History and Guide, (Stroud: Tempus) Hinton, David A. 1977. Alfred’s Kingdom: Wessex and the South (London: J M Dent & Sons) Hooke, Della. 1988. The Landscape of Anglo-Saxon England (London: Leicester University Press) Howell, Isca, and Graham Spurr. 2009. PLANNING DELIVERY ZONE 12, Trench PDZ 12.01, London, E15: A post-excavation assessment, (London: Museum of London Archaeology) Jarman, Catherine L., Martin Biddle, Tom Higham and Christopher Bronk Ramsay. 2018. ‘The Great Viking Army in England: New Dates from the Repton Charnel’, Antiquity, 92 (361): 183-199 Jesch, Judith. 2015. The Viking Diaspora (Abingdon: Routledge) Kershaw, Paul. 2000. ‘The Alfred-Guthrum Treaty: Scripting Accommodation and Interaction in Viking Age England’, in Cultures in Contact: Scandinavian Settlement in the Ninth and Tenth Centuries, ed. by Dawn M. Hadley and Julian D. Richards (Turnhout: Brepols), pp. 43-46 Lavelle, Ryan. 2010. Alfred’s Wars: Sources and Interpretations of Anglo-Saxon Warfare in the Viking Age (Woodbridge: Boydell and Brewer) Lewis-Stempell, John. 2006. England: The Autobiography (London: Penguin) Maull, Anthony and Andy Chapman. 2005. A Medieval Moated Enclosure in Tempsford Park (Great Cransley: Northamptonshire Archaeology Northamptonshire County Council and The Bedfordshire Archaeological Council) McLeod, Shane. 2014. The Beginning of Scandinavian Settlement in England: The Viking ‘Great Army’ and Early Settlers, c. 865-900 (Turnhout: Brepols) Muir, Richard. 2002. The New Reading the Landscape: Fieldwork in Landscape History (Exeter: University of Exeter Press) Ó Corráin, Donnchadh. 2012. ‘The Vikings and Ireland’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 428-433 Olsen, Bjørnar. 2003. ‘Belligerent Chieftains and Oppressed Hunters? Changing Conceptions of Interethnic Relationships in Northern Norway during the Iron Age and Early Medieval Period’, in Contact, Continuity and Collapse: The Norse Colonization of the North Atlantic, ed. by James H. Barrett (Turnhout: Brepols), pp. 9-31 Ordnance Survey. 2001. Historical Map and Guide: Roman Britain, 5th edn, 1: 625,000 (Southampton) Pedersen, Anne. 2012. ‘Viking Weaponry’ in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 204-211 Pitts, Mike. 2012. ‘Digging for the 2012 Olympics’, British Archaeology, 65 (4): 14-23 Price, Neil. 2017. The Viking Way: Magic and Mind in Late Iron Age Scandinavia (Oxford: Oxbow) Renaud, Jean. 2012. ‘The Duchy of Normandy’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 453-457 Reynolds, Andrew. 2002. Later Anglo-Saxon England. Life and Landscape (Stroud: Tempus) Reynolds, Andrew. 2009. Anglo-Saxon Deviant Burial Customs (Oxford: Oxford University Press) Richards, Julian. 2007. Viking Age England (Stroud: Tempus) Rumble, Alexander R. and Mills, Anthony David (ed.). 1997. Names, Places and People: Onomastic Miscellany for John McNeal Dodgson (Stamford: Paul Watkins) Sawyer, Peter. 2007. ‘Ohthere’s Destinations: Norway, Denmark and England’, in Ohthere’s Voyages: A Late 9th-Century Account of Voyages Along the Coasts of Norway and Denmark and its
51 Cultural Context, ed. by Janet Bately and Anton Englert (Roskilde: The Viking Ship Museum) pp. 40-58 Sawyer, Peter. 1968. Anglo-Saxon Charters: An Annotated List and Bibliography (London: Royal Historical Society) Shama, Simon. 2000. A History of Britain: at the Edge of the World? 3000 BC – AD 1603 (London: BBC Worldwide) Stenton, Frank. 1989. Anglo-Saxon England (Oxford: Oxford University Press) Storr, Jim. 2013. ‘Bran, Brent, Fleam and Devil: The Purpose of the Cambridge Dykes’, British Archaeology, 131: 46-49 Swanton, Michael (ed.). 2000. The Anglo-Saxon Chronicles (London: Phoenix Press) Thorpe, Benjamin. 1840. Ancient Laws and Institutes of England, Monumenta Ecclesiastica Anglicana and the Ancient Latin Version of the Anglo-Saxon Laws (London: G.E. Eyre and A. Spottiswoode) Tilly, Charles. 1975. The Formation of National States in Western Europe (Princeton: Princeton University Press) Wallace, Patrick F. 2012. ‘Archaeological Evidence for the Different Expressions of Scandinavian Settlement in Ireland, 840-1100’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 434-438 Whitelock, Dorothy. 1979. English Historical Documents c. 500-1042 (London: Eyre Methueun) Zachrisson, Inger. 2012. ‘The Sámi and their Interaction with the Nordic Peoples’, in The Viking World, ed. by Stefan Brink and Neil Price (Abingdon: Routledge), pp. 32-39
Internet Sources
Anonymous. 2013a. 'English Heritage Pastscapes':
52 Belford, Paul ‘Offa’s Dyke evidence at Chirk suggests earlier build’, BBC Wales (2014),
53